Directory UMM :Data Elmu:jurnal:T:Tree Physiology:Vol14.1994:
Tree Physiology 14, 1351--1366
© 1994 Heron Publishing----Victoria, Canada
Seasonal patterns of light-saturated photosynthesis and leaf
conductance for mature and seedling Quercus rubra L. foliage:
differential sensitivity to ozone exposure
P. J. HANSON,1 L. J. SAMUELSON,2 S. D. WULLSCHLEGER,1
T. A. TABBERER1 and G. S. EDWARDS2
1
Environmental Sciences Division, Oak Ridge National Laboratory, P.O. Box 2008, Oak Ridge,
TN 37831-6034, USA
2
Tennessee Valley Authority, Cooperative Forest Studies Program, TVA Forestry Building, Norris,
TN 37828, USA
Received February 2, 1994
Summary
Extrapolation of the effects of ozone on seedlings to large trees and forest stands is a common objective
of current assessment activities, but few studies have examined whether seedlings are useful surrogates
for understanding how mature trees respond to ozone. This two-year study utilized a replicated open-top
chamber facility to test the effects of subambient, ambient and twice ambient ozone concentrations on
light-saturated net photosynthesis (Pmax) and leaf conductance (gl) of leaves from mature trees and
genetically related seedlings of northern red oak (Quercus rubra L.). Gas exchange measurements were
collected four times during the 1992 and 1993 growing seasons. Both Pmax and gl of all foliage followed
normal seasonal patterns of ontogeny, but mature tree foliage had greater Pmax and gl than seedling foliage
at physiological maturity. At the end of the growing season, Pmax and gl of the mature tree foliage exposed
to ambient (≈80--100 ppm-h) and twice ambient (≈150--190 ppm-h) exposures of ozone were reduced 25
and 50%, respectively, compared with the values for foliage in the subambient ozone treatment
(≈35 ppm-h). In seedling leaves, Pmax and gl were less affected by ozone exposure than in mature leaves.
Extrapolations of the results of seedling exposure studies to foliar responses of mature forests without
considering differences in foliar anatomy and stomatal response between juvenile and mature foliage
may introduce large errors into projections of the response of mature trees to ozone.
Keywords: ozone dose, ozone uptake, air pollutant, foliar anatomy, stomatal response.
Introduction
Photosynthesis and respiration of leaf and stem tissues are key measurements in
stress studies, because of their potential relationship to plant dry matter accumulation. For plant species of small stature and relatively simple canopy structure,
individual leaf responses to manipulated environments have been widely studied, but
it is not clear whether similar measurements with large trees would yield the same
results (Pye 1988, Cregg et al. 1989). The characteristics of large trees that differ
from those of seedlings include increased respiratory mass of structural tissues,
substantial resistances to water and nutrient flux within large stems, modified water
and nutrient availability as a result of greater root exploitation of soils, mutual
shading of foliage within complex canopies, and the presence of ‘‘mature’’ characteristics such as increased carbon demands of support tissue and reproduction.
1352
HANSON ET AL.
Ultimately, it is desirable to generalize measurements taken at the seedling or tissue
level to mature trees and forests, and an understanding of differences between
stress-induced foliar responses of seedlings and those of canopy foliage (if they exist)
is an essential first step in this process.
Observations on an unreplicated study of the impacts of tropospheric ozone on
foliage of mature trees and seedlings of northern red oak (Quercus rubra L.)
indicated that mature tree foliage was more sensitive to external ozone exposure than
seedling foliage, which showed essentially no response (Edwards et al. 1994).
Samuelson and Edwards (1993) similarly reported no impact of ozone on northern
red oak seedling foliage, but found a consistent reduction in foliar photosynthetic
characteristics of mature tree foliage in response to ozone exposure. To obtain more
detailed information about the impacts of tropospheric ozone on northern red oak,
we have analyzed seasonal patterns of responses of mature tree and seedling foliage
to external and internal ozone exposures. Specific objectives were to (i) quantify the
direct effects of variable ozone exposures on seasonal patterns of light-saturated
photosynthesis, leaf conductance and leaf water status for mature tree and seedling
leaves, and (ii) evaluate leaf conductance to ozone as a mechanism explaining
variable ozone sensitivity between tree and seedling foliage.
Methods
Experimental design
The field performance of the experimental facility and the preparation of plant
materials have previously been described by Samuelson and Edwards (1993) and
Edwards et al. (1994). Briefly, nine mature northern red oak trees located in a
30-year-old seed orchard operated by the Tennessee Valley Authority (Norris, TN)
were individually enclosed in large open-top chambers (4.6 × 8.2 m) and exposed to
one of three regimes of ozone exposure: subambient conditions provided by charcoal-filtered air (CF), ambient air (Amb.) or twice the ambient concentration of
ozone (2×). Two-year-old northern red oak seedlings grown from acorns collected in
the seed orchard were transplanted to 24-l pots and placed in standard open-top
chambers (3.0 × 2.4 m) for exposure to the same ozone treatments. The study was
initiated in 1992 with 90 seedlings per treatment (30 per standard chamber) and three
mature trees per treatment (one per chamber). All ozone treatments were initiated in
early April of the 1992 and 1993 growing seasons, and were discontinued prior to
fall senescence (late September). All trees and seedlings were irrigated to maintain
optimum water conditions (>− 0.5 MPa soil matric potential) throughout each growing season.
Foliar measurements
Seasonal patterns of light-saturated photosynthesis (Pmax), leaf conductance to water
vapor (gl) and leaf water potentials were measured during four 2- to 3-day measure-
OZONE SENSITIVITY OF MATURE AND SEEDLING QUERCUS
1353
ment periods in 1992 and 1993. In 1992, measurements were made during the weeks
of May 12, June 16, July 29 and September 15. The 1993 measurements were
conducted during the weeks of May 24, July 7, August 17 and September 17. The
May sampling periods in 1992 and 1993 corresponded to periods of leaf expansion
for both the mature trees and seedlings. For each measurement date, 6--8 leaves from
each mature tree (mid-canopy and south aspect locations) and one leaf from 6--8
seedlings per chamber were evaluated for their light-saturated gas exchange characteristics (Pmax and gl), and an additional pair of leaves was used to assess leaf water
potentials. The seedlings produced more than one flush of leaves within each
growing season, but only the response of the first-flush leaves to ozone exposures
was considered, because the first-flush leaves had the same ozone exposure duration
as the mature tree leaves. All measurements were made between 1000 and 1500 h,
and sampling was conducted by blocks, with each block containing samples from
each of the three treatments.
Both Pmax and gl were measured with a portable photosynthesis system employing
a 1-l chamber (LI-6200, Li-Cor Inc., Lincoln, NE). Artificial lighting supplemented
natural light under partly cloudy conditions to ensure that light-saturated conditions
(PAR > 800 µmol m −2 s −1) were present throughout each 40--50 s sampling interval.
Seedlings were removed from the treatment chambers temporarily for these measurements, and observations were made on attached leaves. The measurements for
mature tree leaves were conducted on detached leaves brought immediately to a
central monitoring location (typically completed within 2 min). Premeasurement
tests showed little change in foliar characteristics for up to 3 min following leaf
detachment. Data were discarded if the sequential 20-s measurements did not agree.
Leaf water potentials were determined for detached seedling and mature tree leaves
with a Scholander pressure cylinder (PMS Instrument Co., Corvallis, OR) based on
established protocols for hardwood foliage (Ritchie and Hinckley 1975, Pallardy et
al. 1991).
Foliage samples were saved after each measurement period of the 1993 growing
season. Foliage area was determined with a Li-Cor leaf area meter, and then the
foliage was freeze-dried before calculating the leaf mass per unit leaf area (g m −2)
for each sample. At the final harvest in September 1993, additional fresh leaf samples
were taken for the assessment of stomatal frequency (number mm −2) and stomatal
size (µm). A portion of the abaxial surface of 4--5 leaves per treatment was coated
with clear nail polish to make an impression of the stomatal features. These impressions were then viewed with transmission lighting on a Nikon SMZ-U stereomicroscope (Nikon Corp., Tokyo, Japan) with a video interface at a combined optical and
electronic magnification near 700×.
Calculations of internal and external ozone exposures
Total cumulative external ozone exposure (24-h Sum00) or dose expressed in ppm-h
(converted from ppb-h for convenience) was determined by summing all mean
hourly ozone concentrations for each treatment from calendar Day 101 through
1354
HANSON ET AL.
Day 260 of each growing season. We also calculated cumulative external ozone
exposures for ozone concentrations above 60 ppb (24-h Sum06) by summing all
mean hourly concentrations throughout the season exceeding the 60 ppb target
concentration. This peak-weighted cumulative index of exposure assumes negligible
ozone effects below the target concentration and has previously been shown to
correlate well with many ozone response variables (Lee et al. 1988).
Internal ozone uptake by foliage (nmol m −2 s −1) was calculated as follows:
[O3] × gl,ozone = Internal O3 deposition,
(1)
where [O3] is the ozone concentration in nmol mol −1, and gl,ozone is the light-saturated
leaf conductance to ozone in mol m −2 s −1 derived from gl values adjusted for the
diffusivity of ozone in air (i.e., multiplied by 0.6; Laisk et al. 1989). This equation
assumes an intercellular ozone concentration close to zero, and it has been documented for crop (Laisk et al. 1989, Neubert et al. 1993) and woody plant (Skärby et
al. 1987, Taylor and Hanson 1992, Wieser and Havraneck 1993) species. Cumulative
ozone uptake (mmol m −2) for the mature and seedling foliage was estimated for each
treatment by integrating Equation 1 for all daylight hourly ozone concentrations from
April 11 (Day 101) through September 17 (Day 260) of 1992 and 1993. Ozone
uptake was assumed to be negligible during dark periods when gl is at a minimum.
Beginning at zero on Day 101, foliar gl values were linearly interpolated between the
four seasonal measurement periods for these calculations (e.g., if gl was measured to
be 0.1 mol m −2 s −1 on Day X and 0.2 mol m −2 s −1 on Day X+10, then it was calculated
to be 0.15 mol m −2 s −1 on Day X+5).
Statistical analyses
The experimental unit was the chamber for both the mature tree and seedling
comparisons yielding three replications per treatment. Within an age grouping (i.e.,
tree versus seedling data) and date, responses of light-saturated photosynthesis, leaf
conductance and leaf water potential to ozone were analyzed by a randomized block
design with blocks and ozone treatments as the main effects. Repeated measures
analyses of these variables were also conducted with ozone treatment as the betweensubjects main effect and seasonal sampling dates represented the repeated trials. May
observations were not used for this analysis because of the predominant influence of
leaf development. Although the repeated measures analysis is somewhat redundant
with the within-date randomized block analyses, it was conducted to determine the
consistency of the ozone response throughout the season (i.e., to test for interactions
between ozone and time). However, the low replication reduced the power of the
repeated measures analysis, making it advantageous to analyze the data within dates
as well. No direct statistical analyses of ozone effects were performed for comparisons between tree and seedling foliar responses because of possible differences
between environments in the standard and large tree exposure chambers, and because
the mean square error differed between age classes (Samuelson and Edwards 1993).
OZONE SENSITIVITY OF MATURE AND SEEDLING QUERCUS
1355
However, we conducted a tree versus seedling analysis to test for differences in leaf
anatomical characteristics in August and September 1993.
Linear regression analyses were used to evaluate the relationship between photosynthetic responses and the ozone exposure indices described previously. The regression analyses were conducted on the combined mean treatment response data from
the 1992 and 1993 growing seasons. Combining the 1992 and 1993 data is appropriate for hardwoods, because the ozone exposures start each year at zero with the new
foliage.
Results
Ozone exposures
The daily 7-h mean monthly ozone concentrations for the 1992 and 1993 seedling
and mature tree treatments are shown in Figure 1. Mean external ozone exposures
for all three treatments in both years were similar for seedling and mature trees. There
was no pronounced seasonal pattern in either year. However, the monthly 7-h mean
ozone concentrations in 1993 were typically higher than those in 1992. For example,
in June 1993, the subambient, ambient and twice ambient treatment concentrations
(averaged across seedling and mature tree chambers) were 28, 46 and 55% greater,
Figure 1. Monthly 7-h mean ozone concentrations (ppb, v/v) averaged across three replicate seedling
(open symbols) and mature tree (closed symbols) chambers for the 1992 and 1993 growing seasons.
Treatments are charcoal-filtered air with subambient ozone (squares), ambient ozone (circles) and twice
ambient ozone (triangles).
1356
HANSON ET AL.
respectively, than the corresponding concentrations in 1992. Mean total cumulative
external ozone exposures (24-h Sum00) were 34, 79 and 147 ppm-h in 1992 and 37,
95 and 188 ppm-h in 1993 for the subambient, ambient and twice ambient treatments,
respectively. Seasonal maximum ambient 1-h mean ozone concentrations in 1992
and 1993 did not exceed the National Ambient Air Quality Standard for ozone
(NAAQS = 120 ppb, US EPA 1986), but the twice ambient treatments exceeded
120 ppb on an average of 20 and 70 days in 1992 and 1993, respectively. Elevated
ambient ozone concentrations in 1993 conveniently provided increased ozone exposures to test relationships between external and internal ozone exposure against foliar
responses.
Foliar responses
Although the mature trees and seedlings were grown in the same soil and maintained
at similar leaf water status (Table 1), Pmax of mature tree foliage at physiological
maturity (mid-June) was twice that of seedling foliage for leaves produced in the
1992 and 1993 growing seasons (Figure 2). Similarly, leaf conductance to water
vapor (gl) of seedling foliage was about half that of the mature tree foliage during the
Table 1. Mean leaf water potential (± 1 SD) for mature tree and first-flush seedling foliage exposed to
charcoal-filtered air (CF), ambient air (Amb.), or air containing twice the ambient ozone concentration
(2×). Significant P-values for ozone treatment main effects < 0.10 are provided after the treatment means
(ns = not significant, nd = no data).
Variable
Leaf water potential (MPa)
1992 Season
Mature tree data
CF
Amb.
2×
P-value
Seedling data
CF
Amb.
2×
P-value
May 11--15
June 15--17
July 28--30
September 14--17
−0.74 ± 0.29
−0.68 ± 0.18
−0.77 ± 0.07
ns
nd
nd
nd
−1.32 ± 0.63
−1.38 ± 0.53
−1.10 ± 0.48
ns
−1.65 ± 0.48
−1.85 ± 0.60
−1.52 ± 0.61
ns
−0.52 ± 0.13
−0.56 ± 0.20
−0.54 ± 0.24
ns
nd
nd
nd
−0.92 ± 0.35
−0.90 ± 0.26
−0.92 ± 0.32
ns
−1.15 ± 0.42
−1.14 ± 0.39
−1.00 ± 0.28
ns
May 24--26
July 7--9
August 17--19
September 17
−1.37 ± 0.15
−1.26 ± 0.37
−1.20 ± 0.29
ns
−1.62 ± 0.37
−1.15 ± 0.35
−1.38 ± 0.20
ns
−1.26 ± 0.66
−1.01 ± 0.12
−0.62 ± 0.22
ns
−1.48 ± 0.16
−1.21 ± 0.36
−0.99 ± 0.26
ns
−1.33 ± 0.21
−1.41 ± 0.11
−1.27 ± 0.20
ns
−1.57 ± 0.37
−1.64 ± 0.41
−1.62 ± 0.30
ns
−0.83 ± 0.39
−0.83 ± 0.39
−0.84 ± 0.38
ns
−0.55 ± 0.18
−0.71 ± 0.27
−0.90 ± 0.33
ns
1993 Season
Mature tree data
CF
Amb.
2×
P-value
Seedling data
CF
Amb.
2×
P-value
OZONE SENSITIVITY OF MATURE AND SEEDLING QUERCUS
1357
Figure 2. Seasonal patterns of light-saturated photosynthesis (Pmax) for seedling leaves (S, open symbols)
and mature tree ‘‘sun’’ foliage (T, closed symbols) exposed to three ozone regimes. Treatments are
charcoal-filtered air with subambient ozone (squares), ambient ozone (circles) and twice ambient ozone
(triangles). Significant differences in Pmax by date (P < 0.05) are indicated by different letters.
early summer (Table 2). Tree leaves had comparable or somewhat more negative
midday water potentials than seedling leaves (Table 1).
In 1992, both mature tree and seedling foliage exhibited increasing Pmax and gl up
to seasonal maxima in June followed by a gradual seasonal decline (Figure 2). The
1993 pattern for mature trees was the same as in 1992, whereas seedling foliage had
essentially constant Pmax throughout 1993, but gl increased from May to early July
(Figure 2 and Table 2).
At the end of the 1993 growing season, seedling leaves showed greater stomatal
frequency than mature tree leaves (Table 3) which contrasts with the pattern for gl
(Table 2). The mean diameter of individual stomatal guard cell pairs, however, was
greater for mature tree leaves than for seedling leaves. Mean leaf mass per area
(LMA, g m −2) of seedling leaves was lower than that of the mature tree leaves during
the 1992 and 1993 growing seasons, but ozone had no effect on leaf anatomical
characteristics for either the mature tree or seedling foliage (Table 3).
In addition to the gradual seasonal decline in Pmax in the subambient ozone
treatments, mature tree foliage also showed significant reductions in Pmax (Figure 1)
and gl (Table 2) with increasing ozone concentrations by the end of each growing
season. Repeated measures analysis of variance indicated a significant ozone response for mature tree foliage for the 1992 (P = 0.008) and 1993 (P = 0.002) growing
seasons. Within-date analyses showed no effect of ozone on first-flush seedling Pmax
1358
HANSON ET AL.
Table 2. Mean leaf conductance to water vapor (gl) (± 1 SD) for mature tree and first-flush seedling
foliage exposed to charcoal-filtered air (CF), ambient air (Amb.), or air containing twice the ambient
ozone concentration (2×). Significant P-values for ozone treatment main effects are provided after the
treatment means (ns = not significant).
Variable
1992 Season
Mature tree data
CF
Amb.
2×
P-value
Seedling data
CF
Amb.
2×
P-value
1993 Season
Mature tree data
CF
Amb.
2×
P-value
Seedling data
CF
Amb.
2×
P-value
Mean gl by date (mol m − 2s − 1)
May 11--15
June 15--17
July 28--30
September 14--17
0.13 ± 0.03
0.12 ± 0.05
0.11 ± 0.03
ns
0.31 ± 0.12
0.24 ± 0.05
0.19 ± 0.12
ns
0.23 ± 0.07
0.24 ± 0.04
0.19 ± 0.05
ns
0.44 ± 0.12
0.35 ± 0.22
0.22 ± 0.12
0.05
0.071 ± 0.01
0.068 ± 0.01
0.066 ± 0.02
ns
0.14 ± 0.02
0.17 ± 0.05
0.14 ± 0.01
ns
0.12 ± 0.02
0.12 ± 0.01
0.11 ± 0.04
ns
0.13 ± 0.07
0.15 ± 0.05
0.10 ± 0.03
ns
May 24--26
July 7--9
August 17--19
September 21
0.10 ± 0.06
0.08 ± 0.03
0.09 ± 0.01
ns
0.34 ± 0.16
0.25 ± 0.02
0.26 ± 0.06
ns
0.27 ± 0.08
0.18 ± 0.06
0.09 ± 0.04
0.08
0.22 ± 0.04
0.14 ± 0.04
0.08 ± 0.03
0.002
0.07 ± 0.02
0.08 ± 0.01
0.08 ± 0.01
ns
0.11 ± 0.04
0.14 ± 0.05
0.12 ± 0.02
ns
0.12 ± 0.04
0.12 ± 0.03
0.13 ± 0.02
ns
0.10 ± 0.03
0.11 ± 0.01
0.12 ± 0.02
ns
(Figure 1) or gl (Table 2) for either the 1992 or 1993 complement of leaves. Repeated
measures analysis of variance for the 1992 seedling leaves showed a trend toward
reduced Pmax with increasing ozone exposures (P = 0.06), but the same analysis for
1993 seedling leaves showed no significant trend (P = 0.17). The data indicate that
seedling Pmax is less sensitive to ozone exposure than Pmax of mature tree foliage.
Response of Pmax to internal and external ozone exposure
Combined 1992 and 1993 Pmax data for the mature tree and seedling foliage expressed as a function of internal ozone exposure (Figure 3A) and cumulative external
exposure indices (Figures 3B and 3C) exhibited negative relationships. The significant negative linear relationships for the mature tree leaves explained approximately
80% of the variability in the data, whereas the seedling regression data showed poor
relationships and explained only 22% of the variability (Table 4). The relationship
between mature tree Pmax and total internal or external (Sum00) exposure explained
76 and 83% of the observed response, respectively. In contrast, the threshold index
(Sum06) explained only 49% of the observed response for the mature tree foliage.
Use of the Sum06 ozone exposure index slightly improved the regression r2 and
OZONE SENSITIVITY OF MATURE AND SEEDLING QUERCUS
1359
Table 3. Mean leaf mass per area (LMA) and foliar stomatal characteristics for mature tree foliage and
seedling leaves exposed to charcoal-filtered air (CF), ambient air (Amb.), or air containing twice the
ambient ozone concentration (2×), and the combined mean value for tree versus seedling foliage. The
P-values for ANOVA treatment main effects greater than 0.10 are listed as not significant (ns). Similar
values were obtained in August 1993.
Variable
September 1993
LMA (g m −2 ± SD)
Mature tree foliage
CF
Amb.
2×
P-value
Seedling foliage
CF
Amb.
2×
P-value
Tree versus seedlings
Tree
Seedling
P-value
84 ± 12
66 ± 4
0.001
Stomatal frequency (number mm −2)
Tree
Seedling
P-value
514 ± 57
666 ± 36
< 0.001
Diameter of stomatal guard cell pair (µm)
Tree
Seedling
P-value
27 ± 1
22 ± 1
< 0.001
88 ± 10
73 ± 5
92 ± 11
ns
69 ± 1
66 ± 4
64 ± 4
ns
P-value for the seedling Pmax data.
The negative relationships in Figure 3 include a component of seasonal decline due
to ‘‘natural’’ aging processes (Hanson et al. 1988, Heichel and Turner 1983). To
remove the age influence from the relationship between Pmax and internal/external
ozone exposure, the periodic seedling and mature tree Pmax values from Figure 3 were
expressed relative to their respective mean Pmax values in charcoal-filtered air
(Figure 4). These plots allow a more direct comparison of mature tree and seedling
responses relative to their respective absolute Pmax values, and they can be interpreted
as a fractional reduction in Pmax values over time in response to ozone exposure.
These normalized data retain the negative relationships between foliar photosynthetic capacity and cumulative internal or external exposures, and indicate separate
relationships for seedlings and mature tree Pmax as a function of cumulative external
exposure indices (Figures 4B and 4C). Based on the tests recommended by Draper
and Smith (1981), we determined that the mature tree and seedling regressions for
relative Pmax (Figure 4) were significantly different (P < 0.05) for all three exposure
indices. Both ozone uptake and total cumulative external exposure (Sum00) ex-
1360
HANSON ET AL.
Figure 3. Light-saturated photosynthesis (Pmax) of mature tree (j) and seedling (s) leaves for combined
data from the 1992 and 1993 growing seasons shown as a function of internal ozone uptake (A), total
cumulative external ozone exposure (B, 24-h Sum00), and cumulative exposure to concentrations
> 60 ppb (C, 24-h Sum06). Coefficients for the linear regressions are provided in Table 4.
plained a large percentage (70%) of the mature tree relative Pmax response. Expressing the external exposure with the Sum06 did not improve the linear fit to the relative
Pmax response for the mature tree foliage (Figure 4, Table 4), but the regression of
seedling relative Pmax data against the Sum06 index showed some improvement over
the Sum00 index (r2 = 0.36 and 0.28, respectively; Table 4).
Discussion
Mature versus seedling foliar characteristics
In both mature tree and seedling foliage, Pmax increased to a maximum in June at the
time of physiological maturity (Figure 2) and then declined gradually throughout the
remainder of the growing season. The net photosynthetic rates and leaf conductances
documented for the mature tree leaves near physiological maturity (June data) were
OZONE SENSITIVITY OF MATURE AND SEEDLING QUERCUS
1361
Table 4. Regression coefficients (± SE) for the linear regressions in Figures 3 and 4 relating foliar
response to ozone external (dose) or internal exposures (uptake). Significant probabilities (P-values)
following the regression coefficients reflect a rejection of the null hypothesis that the slope of the linear
regression is equal to zero.
Relationships
Pmax versus
Uptake
Tree
Seedling
Sum00
Tree
Seedling
Sum06
Tree
Seedling
Relative Pmax versus
Uptake
Tree
Seedling
Sum00
Tree
Seedling
Sum06
Tree
Seedling
Linear coefficients
Intercept
Slope
r2
P-value
17.6 ± 0.8
8.5 ± 0.5
−0.375 ± 0.043
−0.071 ± 0.043
0.83
0.15
© 1994 Heron Publishing----Victoria, Canada
Seasonal patterns of light-saturated photosynthesis and leaf
conductance for mature and seedling Quercus rubra L. foliage:
differential sensitivity to ozone exposure
P. J. HANSON,1 L. J. SAMUELSON,2 S. D. WULLSCHLEGER,1
T. A. TABBERER1 and G. S. EDWARDS2
1
Environmental Sciences Division, Oak Ridge National Laboratory, P.O. Box 2008, Oak Ridge,
TN 37831-6034, USA
2
Tennessee Valley Authority, Cooperative Forest Studies Program, TVA Forestry Building, Norris,
TN 37828, USA
Received February 2, 1994
Summary
Extrapolation of the effects of ozone on seedlings to large trees and forest stands is a common objective
of current assessment activities, but few studies have examined whether seedlings are useful surrogates
for understanding how mature trees respond to ozone. This two-year study utilized a replicated open-top
chamber facility to test the effects of subambient, ambient and twice ambient ozone concentrations on
light-saturated net photosynthesis (Pmax) and leaf conductance (gl) of leaves from mature trees and
genetically related seedlings of northern red oak (Quercus rubra L.). Gas exchange measurements were
collected four times during the 1992 and 1993 growing seasons. Both Pmax and gl of all foliage followed
normal seasonal patterns of ontogeny, but mature tree foliage had greater Pmax and gl than seedling foliage
at physiological maturity. At the end of the growing season, Pmax and gl of the mature tree foliage exposed
to ambient (≈80--100 ppm-h) and twice ambient (≈150--190 ppm-h) exposures of ozone were reduced 25
and 50%, respectively, compared with the values for foliage in the subambient ozone treatment
(≈35 ppm-h). In seedling leaves, Pmax and gl were less affected by ozone exposure than in mature leaves.
Extrapolations of the results of seedling exposure studies to foliar responses of mature forests without
considering differences in foliar anatomy and stomatal response between juvenile and mature foliage
may introduce large errors into projections of the response of mature trees to ozone.
Keywords: ozone dose, ozone uptake, air pollutant, foliar anatomy, stomatal response.
Introduction
Photosynthesis and respiration of leaf and stem tissues are key measurements in
stress studies, because of their potential relationship to plant dry matter accumulation. For plant species of small stature and relatively simple canopy structure,
individual leaf responses to manipulated environments have been widely studied, but
it is not clear whether similar measurements with large trees would yield the same
results (Pye 1988, Cregg et al. 1989). The characteristics of large trees that differ
from those of seedlings include increased respiratory mass of structural tissues,
substantial resistances to water and nutrient flux within large stems, modified water
and nutrient availability as a result of greater root exploitation of soils, mutual
shading of foliage within complex canopies, and the presence of ‘‘mature’’ characteristics such as increased carbon demands of support tissue and reproduction.
1352
HANSON ET AL.
Ultimately, it is desirable to generalize measurements taken at the seedling or tissue
level to mature trees and forests, and an understanding of differences between
stress-induced foliar responses of seedlings and those of canopy foliage (if they exist)
is an essential first step in this process.
Observations on an unreplicated study of the impacts of tropospheric ozone on
foliage of mature trees and seedlings of northern red oak (Quercus rubra L.)
indicated that mature tree foliage was more sensitive to external ozone exposure than
seedling foliage, which showed essentially no response (Edwards et al. 1994).
Samuelson and Edwards (1993) similarly reported no impact of ozone on northern
red oak seedling foliage, but found a consistent reduction in foliar photosynthetic
characteristics of mature tree foliage in response to ozone exposure. To obtain more
detailed information about the impacts of tropospheric ozone on northern red oak,
we have analyzed seasonal patterns of responses of mature tree and seedling foliage
to external and internal ozone exposures. Specific objectives were to (i) quantify the
direct effects of variable ozone exposures on seasonal patterns of light-saturated
photosynthesis, leaf conductance and leaf water status for mature tree and seedling
leaves, and (ii) evaluate leaf conductance to ozone as a mechanism explaining
variable ozone sensitivity between tree and seedling foliage.
Methods
Experimental design
The field performance of the experimental facility and the preparation of plant
materials have previously been described by Samuelson and Edwards (1993) and
Edwards et al. (1994). Briefly, nine mature northern red oak trees located in a
30-year-old seed orchard operated by the Tennessee Valley Authority (Norris, TN)
were individually enclosed in large open-top chambers (4.6 × 8.2 m) and exposed to
one of three regimes of ozone exposure: subambient conditions provided by charcoal-filtered air (CF), ambient air (Amb.) or twice the ambient concentration of
ozone (2×). Two-year-old northern red oak seedlings grown from acorns collected in
the seed orchard were transplanted to 24-l pots and placed in standard open-top
chambers (3.0 × 2.4 m) for exposure to the same ozone treatments. The study was
initiated in 1992 with 90 seedlings per treatment (30 per standard chamber) and three
mature trees per treatment (one per chamber). All ozone treatments were initiated in
early April of the 1992 and 1993 growing seasons, and were discontinued prior to
fall senescence (late September). All trees and seedlings were irrigated to maintain
optimum water conditions (>− 0.5 MPa soil matric potential) throughout each growing season.
Foliar measurements
Seasonal patterns of light-saturated photosynthesis (Pmax), leaf conductance to water
vapor (gl) and leaf water potentials were measured during four 2- to 3-day measure-
OZONE SENSITIVITY OF MATURE AND SEEDLING QUERCUS
1353
ment periods in 1992 and 1993. In 1992, measurements were made during the weeks
of May 12, June 16, July 29 and September 15. The 1993 measurements were
conducted during the weeks of May 24, July 7, August 17 and September 17. The
May sampling periods in 1992 and 1993 corresponded to periods of leaf expansion
for both the mature trees and seedlings. For each measurement date, 6--8 leaves from
each mature tree (mid-canopy and south aspect locations) and one leaf from 6--8
seedlings per chamber were evaluated for their light-saturated gas exchange characteristics (Pmax and gl), and an additional pair of leaves was used to assess leaf water
potentials. The seedlings produced more than one flush of leaves within each
growing season, but only the response of the first-flush leaves to ozone exposures
was considered, because the first-flush leaves had the same ozone exposure duration
as the mature tree leaves. All measurements were made between 1000 and 1500 h,
and sampling was conducted by blocks, with each block containing samples from
each of the three treatments.
Both Pmax and gl were measured with a portable photosynthesis system employing
a 1-l chamber (LI-6200, Li-Cor Inc., Lincoln, NE). Artificial lighting supplemented
natural light under partly cloudy conditions to ensure that light-saturated conditions
(PAR > 800 µmol m −2 s −1) were present throughout each 40--50 s sampling interval.
Seedlings were removed from the treatment chambers temporarily for these measurements, and observations were made on attached leaves. The measurements for
mature tree leaves were conducted on detached leaves brought immediately to a
central monitoring location (typically completed within 2 min). Premeasurement
tests showed little change in foliar characteristics for up to 3 min following leaf
detachment. Data were discarded if the sequential 20-s measurements did not agree.
Leaf water potentials were determined for detached seedling and mature tree leaves
with a Scholander pressure cylinder (PMS Instrument Co., Corvallis, OR) based on
established protocols for hardwood foliage (Ritchie and Hinckley 1975, Pallardy et
al. 1991).
Foliage samples were saved after each measurement period of the 1993 growing
season. Foliage area was determined with a Li-Cor leaf area meter, and then the
foliage was freeze-dried before calculating the leaf mass per unit leaf area (g m −2)
for each sample. At the final harvest in September 1993, additional fresh leaf samples
were taken for the assessment of stomatal frequency (number mm −2) and stomatal
size (µm). A portion of the abaxial surface of 4--5 leaves per treatment was coated
with clear nail polish to make an impression of the stomatal features. These impressions were then viewed with transmission lighting on a Nikon SMZ-U stereomicroscope (Nikon Corp., Tokyo, Japan) with a video interface at a combined optical and
electronic magnification near 700×.
Calculations of internal and external ozone exposures
Total cumulative external ozone exposure (24-h Sum00) or dose expressed in ppm-h
(converted from ppb-h for convenience) was determined by summing all mean
hourly ozone concentrations for each treatment from calendar Day 101 through
1354
HANSON ET AL.
Day 260 of each growing season. We also calculated cumulative external ozone
exposures for ozone concentrations above 60 ppb (24-h Sum06) by summing all
mean hourly concentrations throughout the season exceeding the 60 ppb target
concentration. This peak-weighted cumulative index of exposure assumes negligible
ozone effects below the target concentration and has previously been shown to
correlate well with many ozone response variables (Lee et al. 1988).
Internal ozone uptake by foliage (nmol m −2 s −1) was calculated as follows:
[O3] × gl,ozone = Internal O3 deposition,
(1)
where [O3] is the ozone concentration in nmol mol −1, and gl,ozone is the light-saturated
leaf conductance to ozone in mol m −2 s −1 derived from gl values adjusted for the
diffusivity of ozone in air (i.e., multiplied by 0.6; Laisk et al. 1989). This equation
assumes an intercellular ozone concentration close to zero, and it has been documented for crop (Laisk et al. 1989, Neubert et al. 1993) and woody plant (Skärby et
al. 1987, Taylor and Hanson 1992, Wieser and Havraneck 1993) species. Cumulative
ozone uptake (mmol m −2) for the mature and seedling foliage was estimated for each
treatment by integrating Equation 1 for all daylight hourly ozone concentrations from
April 11 (Day 101) through September 17 (Day 260) of 1992 and 1993. Ozone
uptake was assumed to be negligible during dark periods when gl is at a minimum.
Beginning at zero on Day 101, foliar gl values were linearly interpolated between the
four seasonal measurement periods for these calculations (e.g., if gl was measured to
be 0.1 mol m −2 s −1 on Day X and 0.2 mol m −2 s −1 on Day X+10, then it was calculated
to be 0.15 mol m −2 s −1 on Day X+5).
Statistical analyses
The experimental unit was the chamber for both the mature tree and seedling
comparisons yielding three replications per treatment. Within an age grouping (i.e.,
tree versus seedling data) and date, responses of light-saturated photosynthesis, leaf
conductance and leaf water potential to ozone were analyzed by a randomized block
design with blocks and ozone treatments as the main effects. Repeated measures
analyses of these variables were also conducted with ozone treatment as the betweensubjects main effect and seasonal sampling dates represented the repeated trials. May
observations were not used for this analysis because of the predominant influence of
leaf development. Although the repeated measures analysis is somewhat redundant
with the within-date randomized block analyses, it was conducted to determine the
consistency of the ozone response throughout the season (i.e., to test for interactions
between ozone and time). However, the low replication reduced the power of the
repeated measures analysis, making it advantageous to analyze the data within dates
as well. No direct statistical analyses of ozone effects were performed for comparisons between tree and seedling foliar responses because of possible differences
between environments in the standard and large tree exposure chambers, and because
the mean square error differed between age classes (Samuelson and Edwards 1993).
OZONE SENSITIVITY OF MATURE AND SEEDLING QUERCUS
1355
However, we conducted a tree versus seedling analysis to test for differences in leaf
anatomical characteristics in August and September 1993.
Linear regression analyses were used to evaluate the relationship between photosynthetic responses and the ozone exposure indices described previously. The regression analyses were conducted on the combined mean treatment response data from
the 1992 and 1993 growing seasons. Combining the 1992 and 1993 data is appropriate for hardwoods, because the ozone exposures start each year at zero with the new
foliage.
Results
Ozone exposures
The daily 7-h mean monthly ozone concentrations for the 1992 and 1993 seedling
and mature tree treatments are shown in Figure 1. Mean external ozone exposures
for all three treatments in both years were similar for seedling and mature trees. There
was no pronounced seasonal pattern in either year. However, the monthly 7-h mean
ozone concentrations in 1993 were typically higher than those in 1992. For example,
in June 1993, the subambient, ambient and twice ambient treatment concentrations
(averaged across seedling and mature tree chambers) were 28, 46 and 55% greater,
Figure 1. Monthly 7-h mean ozone concentrations (ppb, v/v) averaged across three replicate seedling
(open symbols) and mature tree (closed symbols) chambers for the 1992 and 1993 growing seasons.
Treatments are charcoal-filtered air with subambient ozone (squares), ambient ozone (circles) and twice
ambient ozone (triangles).
1356
HANSON ET AL.
respectively, than the corresponding concentrations in 1992. Mean total cumulative
external ozone exposures (24-h Sum00) were 34, 79 and 147 ppm-h in 1992 and 37,
95 and 188 ppm-h in 1993 for the subambient, ambient and twice ambient treatments,
respectively. Seasonal maximum ambient 1-h mean ozone concentrations in 1992
and 1993 did not exceed the National Ambient Air Quality Standard for ozone
(NAAQS = 120 ppb, US EPA 1986), but the twice ambient treatments exceeded
120 ppb on an average of 20 and 70 days in 1992 and 1993, respectively. Elevated
ambient ozone concentrations in 1993 conveniently provided increased ozone exposures to test relationships between external and internal ozone exposure against foliar
responses.
Foliar responses
Although the mature trees and seedlings were grown in the same soil and maintained
at similar leaf water status (Table 1), Pmax of mature tree foliage at physiological
maturity (mid-June) was twice that of seedling foliage for leaves produced in the
1992 and 1993 growing seasons (Figure 2). Similarly, leaf conductance to water
vapor (gl) of seedling foliage was about half that of the mature tree foliage during the
Table 1. Mean leaf water potential (± 1 SD) for mature tree and first-flush seedling foliage exposed to
charcoal-filtered air (CF), ambient air (Amb.), or air containing twice the ambient ozone concentration
(2×). Significant P-values for ozone treatment main effects < 0.10 are provided after the treatment means
(ns = not significant, nd = no data).
Variable
Leaf water potential (MPa)
1992 Season
Mature tree data
CF
Amb.
2×
P-value
Seedling data
CF
Amb.
2×
P-value
May 11--15
June 15--17
July 28--30
September 14--17
−0.74 ± 0.29
−0.68 ± 0.18
−0.77 ± 0.07
ns
nd
nd
nd
−1.32 ± 0.63
−1.38 ± 0.53
−1.10 ± 0.48
ns
−1.65 ± 0.48
−1.85 ± 0.60
−1.52 ± 0.61
ns
−0.52 ± 0.13
−0.56 ± 0.20
−0.54 ± 0.24
ns
nd
nd
nd
−0.92 ± 0.35
−0.90 ± 0.26
−0.92 ± 0.32
ns
−1.15 ± 0.42
−1.14 ± 0.39
−1.00 ± 0.28
ns
May 24--26
July 7--9
August 17--19
September 17
−1.37 ± 0.15
−1.26 ± 0.37
−1.20 ± 0.29
ns
−1.62 ± 0.37
−1.15 ± 0.35
−1.38 ± 0.20
ns
−1.26 ± 0.66
−1.01 ± 0.12
−0.62 ± 0.22
ns
−1.48 ± 0.16
−1.21 ± 0.36
−0.99 ± 0.26
ns
−1.33 ± 0.21
−1.41 ± 0.11
−1.27 ± 0.20
ns
−1.57 ± 0.37
−1.64 ± 0.41
−1.62 ± 0.30
ns
−0.83 ± 0.39
−0.83 ± 0.39
−0.84 ± 0.38
ns
−0.55 ± 0.18
−0.71 ± 0.27
−0.90 ± 0.33
ns
1993 Season
Mature tree data
CF
Amb.
2×
P-value
Seedling data
CF
Amb.
2×
P-value
OZONE SENSITIVITY OF MATURE AND SEEDLING QUERCUS
1357
Figure 2. Seasonal patterns of light-saturated photosynthesis (Pmax) for seedling leaves (S, open symbols)
and mature tree ‘‘sun’’ foliage (T, closed symbols) exposed to three ozone regimes. Treatments are
charcoal-filtered air with subambient ozone (squares), ambient ozone (circles) and twice ambient ozone
(triangles). Significant differences in Pmax by date (P < 0.05) are indicated by different letters.
early summer (Table 2). Tree leaves had comparable or somewhat more negative
midday water potentials than seedling leaves (Table 1).
In 1992, both mature tree and seedling foliage exhibited increasing Pmax and gl up
to seasonal maxima in June followed by a gradual seasonal decline (Figure 2). The
1993 pattern for mature trees was the same as in 1992, whereas seedling foliage had
essentially constant Pmax throughout 1993, but gl increased from May to early July
(Figure 2 and Table 2).
At the end of the 1993 growing season, seedling leaves showed greater stomatal
frequency than mature tree leaves (Table 3) which contrasts with the pattern for gl
(Table 2). The mean diameter of individual stomatal guard cell pairs, however, was
greater for mature tree leaves than for seedling leaves. Mean leaf mass per area
(LMA, g m −2) of seedling leaves was lower than that of the mature tree leaves during
the 1992 and 1993 growing seasons, but ozone had no effect on leaf anatomical
characteristics for either the mature tree or seedling foliage (Table 3).
In addition to the gradual seasonal decline in Pmax in the subambient ozone
treatments, mature tree foliage also showed significant reductions in Pmax (Figure 1)
and gl (Table 2) with increasing ozone concentrations by the end of each growing
season. Repeated measures analysis of variance indicated a significant ozone response for mature tree foliage for the 1992 (P = 0.008) and 1993 (P = 0.002) growing
seasons. Within-date analyses showed no effect of ozone on first-flush seedling Pmax
1358
HANSON ET AL.
Table 2. Mean leaf conductance to water vapor (gl) (± 1 SD) for mature tree and first-flush seedling
foliage exposed to charcoal-filtered air (CF), ambient air (Amb.), or air containing twice the ambient
ozone concentration (2×). Significant P-values for ozone treatment main effects are provided after the
treatment means (ns = not significant).
Variable
1992 Season
Mature tree data
CF
Amb.
2×
P-value
Seedling data
CF
Amb.
2×
P-value
1993 Season
Mature tree data
CF
Amb.
2×
P-value
Seedling data
CF
Amb.
2×
P-value
Mean gl by date (mol m − 2s − 1)
May 11--15
June 15--17
July 28--30
September 14--17
0.13 ± 0.03
0.12 ± 0.05
0.11 ± 0.03
ns
0.31 ± 0.12
0.24 ± 0.05
0.19 ± 0.12
ns
0.23 ± 0.07
0.24 ± 0.04
0.19 ± 0.05
ns
0.44 ± 0.12
0.35 ± 0.22
0.22 ± 0.12
0.05
0.071 ± 0.01
0.068 ± 0.01
0.066 ± 0.02
ns
0.14 ± 0.02
0.17 ± 0.05
0.14 ± 0.01
ns
0.12 ± 0.02
0.12 ± 0.01
0.11 ± 0.04
ns
0.13 ± 0.07
0.15 ± 0.05
0.10 ± 0.03
ns
May 24--26
July 7--9
August 17--19
September 21
0.10 ± 0.06
0.08 ± 0.03
0.09 ± 0.01
ns
0.34 ± 0.16
0.25 ± 0.02
0.26 ± 0.06
ns
0.27 ± 0.08
0.18 ± 0.06
0.09 ± 0.04
0.08
0.22 ± 0.04
0.14 ± 0.04
0.08 ± 0.03
0.002
0.07 ± 0.02
0.08 ± 0.01
0.08 ± 0.01
ns
0.11 ± 0.04
0.14 ± 0.05
0.12 ± 0.02
ns
0.12 ± 0.04
0.12 ± 0.03
0.13 ± 0.02
ns
0.10 ± 0.03
0.11 ± 0.01
0.12 ± 0.02
ns
(Figure 1) or gl (Table 2) for either the 1992 or 1993 complement of leaves. Repeated
measures analysis of variance for the 1992 seedling leaves showed a trend toward
reduced Pmax with increasing ozone exposures (P = 0.06), but the same analysis for
1993 seedling leaves showed no significant trend (P = 0.17). The data indicate that
seedling Pmax is less sensitive to ozone exposure than Pmax of mature tree foliage.
Response of Pmax to internal and external ozone exposure
Combined 1992 and 1993 Pmax data for the mature tree and seedling foliage expressed as a function of internal ozone exposure (Figure 3A) and cumulative external
exposure indices (Figures 3B and 3C) exhibited negative relationships. The significant negative linear relationships for the mature tree leaves explained approximately
80% of the variability in the data, whereas the seedling regression data showed poor
relationships and explained only 22% of the variability (Table 4). The relationship
between mature tree Pmax and total internal or external (Sum00) exposure explained
76 and 83% of the observed response, respectively. In contrast, the threshold index
(Sum06) explained only 49% of the observed response for the mature tree foliage.
Use of the Sum06 ozone exposure index slightly improved the regression r2 and
OZONE SENSITIVITY OF MATURE AND SEEDLING QUERCUS
1359
Table 3. Mean leaf mass per area (LMA) and foliar stomatal characteristics for mature tree foliage and
seedling leaves exposed to charcoal-filtered air (CF), ambient air (Amb.), or air containing twice the
ambient ozone concentration (2×), and the combined mean value for tree versus seedling foliage. The
P-values for ANOVA treatment main effects greater than 0.10 are listed as not significant (ns). Similar
values were obtained in August 1993.
Variable
September 1993
LMA (g m −2 ± SD)
Mature tree foliage
CF
Amb.
2×
P-value
Seedling foliage
CF
Amb.
2×
P-value
Tree versus seedlings
Tree
Seedling
P-value
84 ± 12
66 ± 4
0.001
Stomatal frequency (number mm −2)
Tree
Seedling
P-value
514 ± 57
666 ± 36
< 0.001
Diameter of stomatal guard cell pair (µm)
Tree
Seedling
P-value
27 ± 1
22 ± 1
< 0.001
88 ± 10
73 ± 5
92 ± 11
ns
69 ± 1
66 ± 4
64 ± 4
ns
P-value for the seedling Pmax data.
The negative relationships in Figure 3 include a component of seasonal decline due
to ‘‘natural’’ aging processes (Hanson et al. 1988, Heichel and Turner 1983). To
remove the age influence from the relationship between Pmax and internal/external
ozone exposure, the periodic seedling and mature tree Pmax values from Figure 3 were
expressed relative to their respective mean Pmax values in charcoal-filtered air
(Figure 4). These plots allow a more direct comparison of mature tree and seedling
responses relative to their respective absolute Pmax values, and they can be interpreted
as a fractional reduction in Pmax values over time in response to ozone exposure.
These normalized data retain the negative relationships between foliar photosynthetic capacity and cumulative internal or external exposures, and indicate separate
relationships for seedlings and mature tree Pmax as a function of cumulative external
exposure indices (Figures 4B and 4C). Based on the tests recommended by Draper
and Smith (1981), we determined that the mature tree and seedling regressions for
relative Pmax (Figure 4) were significantly different (P < 0.05) for all three exposure
indices. Both ozone uptake and total cumulative external exposure (Sum00) ex-
1360
HANSON ET AL.
Figure 3. Light-saturated photosynthesis (Pmax) of mature tree (j) and seedling (s) leaves for combined
data from the 1992 and 1993 growing seasons shown as a function of internal ozone uptake (A), total
cumulative external ozone exposure (B, 24-h Sum00), and cumulative exposure to concentrations
> 60 ppb (C, 24-h Sum06). Coefficients for the linear regressions are provided in Table 4.
plained a large percentage (70%) of the mature tree relative Pmax response. Expressing the external exposure with the Sum06 did not improve the linear fit to the relative
Pmax response for the mature tree foliage (Figure 4, Table 4), but the regression of
seedling relative Pmax data against the Sum06 index showed some improvement over
the Sum00 index (r2 = 0.36 and 0.28, respectively; Table 4).
Discussion
Mature versus seedling foliar characteristics
In both mature tree and seedling foliage, Pmax increased to a maximum in June at the
time of physiological maturity (Figure 2) and then declined gradually throughout the
remainder of the growing season. The net photosynthetic rates and leaf conductances
documented for the mature tree leaves near physiological maturity (June data) were
OZONE SENSITIVITY OF MATURE AND SEEDLING QUERCUS
1361
Table 4. Regression coefficients (± SE) for the linear regressions in Figures 3 and 4 relating foliar
response to ozone external (dose) or internal exposures (uptake). Significant probabilities (P-values)
following the regression coefficients reflect a rejection of the null hypothesis that the slope of the linear
regression is equal to zero.
Relationships
Pmax versus
Uptake
Tree
Seedling
Sum00
Tree
Seedling
Sum06
Tree
Seedling
Relative Pmax versus
Uptake
Tree
Seedling
Sum00
Tree
Seedling
Sum06
Tree
Seedling
Linear coefficients
Intercept
Slope
r2
P-value
17.6 ± 0.8
8.5 ± 0.5
−0.375 ± 0.043
−0.071 ± 0.043
0.83
0.15